Run AI Workloads and Microservices on Kubernetes with Volcano

Explore the capabilities and advancements of Volcano, a cloud-native batch system and the first batch computing project in CNCF, in this informative conference talk. Discover how Volcano enhances performance for high-performance computing workloads in big data, AI, and gene computing. Learn about its job-based fair-share, priority, preemption, reclaim, and queue management features, as well as its integration with computing ecosystems like spark-operator, flink-operator, kubeflow, Ray, and Cromwell. Gain insights into Volcano's recent achievements, including new features for optimizing resource utilization in large-scale clusters, co-location, and resource-aware scheduling enhancements for AIGC. Understand the project's growing adoption, with over 50 users deploying Volcano in production environments. Get an overview of the latest progress, new sub-projects, and the future direction of the Volcano community.

Run Your AI Workloads and Microservices on Kubernetes More Easily and Efficiently... - William Wang